2.1. Differentiating between pandillas and maras

1Youth gangs have existed since the 1960s and 1970s in Central America. However, there are different types of Central American gangs and thus one has to distinguish between pandillas and maras.1 The former are localized, homegrown gangs, which are “direct inheritors” (Jütersonke, Rodgers & Muggah 2009: 379) of the gangs that have historically characterized Central American societies, while the latter are a more recent phenomenon with transnational origins (Ribando Seelke 2011: 4). During the civil war period, pandillas were to be found all over Central America, whereas now they are primarily visible in Nicaragua and to a lesser extent in Costa Rica (where they are known as chapulines) (Jütersonke et al. 2009: 379). In the northern triangle of Central America – El Salvador, Honduras and Guatemala – pandillas were almost entirely replaced by maras. Contemporary pandillas have their roots in the post-conflict period of local socio-economic uncertainty and insecurity, during which returning demobilized combatants banded together. In comparison to previous types, post-conflict pandillas were more violent (partially due to military skills gained during the civil conflicts) and more institutionalized (Ibid.).2

2Maras, on the other hand, do not have local but transnational roots that are closely linked to migratory patterns. Guatemala and El Salvador experienced the longest and second-longest civil wars in Latin American history, from 1960 to 1996 and from 1980 to 1992 respectively, conflicts that caused thousands of Central Americans to leave their home countries, migrating up north, especially (but not exclusively) to Los Angeles in the United States. It was primarily in this environment that the two main gangs, the Mara Salvatrucha (MS-13) and Mara Dieciocho or 18^th Street Gang (M-18), emerged and grew in size.3 The M-18 was initially founded by Mexican immigrants in the 1960s and grew enormously due to arriving migrants from El Salvador and Guatemala. Becoming a gang member was a way for the Central American refugees to “feel included as outsiders in the United States” (Rodgers, Muggah & Stevenson 2009: 7). The Mara Salvatrucha was created in the second half of the 1980s by another wave of Salvadoran migrants in order to compete with the previously existing M-18 (Bruneau 2005: 2), and the two gangs quickly became “bitter rivals” (Jütersonke et al. 2009: 380). With the end of the Salvadoran and Guatemalan civil wars, several gang members returned to Central America on a voluntary basis. As a consequence of more restrictive immigration laws, such as the Illegal Immigration Reform and Immigrant Responsibility Act (IIRIRA) passed in 1996, the U.S. government additionally deported tens of thousands.4

3Upon their return in Central America, the deportees started replicating gang structures and practices that they had become familiar with in the United States by establishing local chapters of the maras, the so-called clikas, in their communities of origin (Jütersonke et al. 2009: 380).5 Especially in Honduras and El Salvador, local pandillas turned into clikas of the MS-13 and M-18, whereas the maras did not absorb the local Guatemalan gangs in this way. The majority of the currently active youth gangs in Guatemala have their roots in the country’s local gangs that appeared in the 1980s (Ranum 2011: 73). The Guatemalan gang phenomenon did not purely originate as a result of deported gang members from the United States; “there are multifaceted reasons for the increase in gang violence and […] gangs have evolved and transformed within the Guatemalan environment” (Ibid.; emphasis added). Nonetheless, the arrival of members of the Californian gangs had an important influence on the Guatemalan pandillas (Savenije 2007). In this way, U.S. gang culture was transnationally transposed to the countries of the Central American northern triangle, which highlights an additional difference from the pandillas, for example, in Nicaragua (Rodgers et al. 2009: 9).6

2.2. Three generations of gangs

4Besides broadly differentiating between pandillas and maras, several studies have classified gangs according to different generations: the first generation (localized, turf-oriented, non-sophisticated groups with a loose leadership structure), the second generation (market-oriented, drug-centered national groups with some transnational links) and the third generation (highly sophisticated, financially-oriented groups that operate internationally) (Franco 2008: 5f; Sullivan 2008; Ribando Seelke 2011: 4). The Mara Salvatrucha and the Mara Dieciocho are commonly considered to meet the characteristic of being active in more than one country (Franco 2008: 8). Building on evidence about contact and communication between different chapters of the maras across Central America, Mexico, the U.S. and even Canada, security specialists suggest that these gangs have the potential to transform from street gangs (first and second generation) into more elaborate transnational criminal groups with a higher degree of organization in their criminal activities (third generation).7 Going beyond this, Cruz (2010: 379) argues that the maras have already evolved into “protection rackets with features of transnational organisations” – a process boosted by repressive policies on the part of the governments of the northern triangle.

5With the end of the Cold War, international relations became subject to changing trends in the security field, for example a shift of attention from traditional security threats to those caused by transnational crime (Edwards & Gill 2002; Giraldo & Trinkunas 2010). Higher cross-border mobility is often named as one of the main reasons for increased emphasis on transnational crime after the collapse of the Soviet Union. As regards the Central American case, transnational organized crime became a greatly debated topic from the 1990s onwards – not only due to the changing security environment after the Cold War but also because of the end of the civil wars in the region. The above-outlined migratory background and transnational roots of the maras is one crucial factor in this debate. Furthermore, the maras underwent a transformation in the 1990s, which was accompanied by a significant increase in criminal activities and gang-related violence (Rodgers et al. 2009; Gutiérrez Rivera 2010). Despite their migratory background, Central American maras not only emerged as an external problem due to deportees from the United States, as at the same time these gangs are deeply rooted in domestic conditions. These include cultures of violence, rapid urban growth, dysfunctional families, socio-economic inequalities as well as a lack of educational and employment opportunities, and the gangs have developed as a consequence (Hume 2007: 744; Rodgers et al. 2009: 9; Brands 2010: 25; Ranum 2011: Ribando Seelke 2011: 6).

6Western policy-makers, security specialists, academics and the media have perceived the links between gangs and criminal organizations or transnational organized crime (TOC) differently. The media in particular depict Central American gangs in “numerous sensationalist accounts” as being linked to “migrant trafficking, kidnapping, and international organized crime” (Rodgers et al. 2009: 9). In contrast to this and based on qualitative studies, Rodgers et al. state that “both pandillas and maras are mainly involved in small-scale, localized crime and delinquency, such as petty theft and muggings (although these can often result in murder)” (Ibid.).8 While such activities are usually carried out interpersonally, the maras in the northern triangle become more and more engaged in collective endeavors (Jütersonke et al. 2009: 381).9 In addition, Jütersonke et al. (2009: 380) hold that “neither gang is a real federal structure, much less a transnational one”. Bernardo Arévalo de León affirms this when saying that Central American maras do not carry out transborder operations, which is why they do not need centralized structures of authority.10 Furthermore, he refers to maras more as identity bands rather than groups involved in economically profitable activities such as the drug trade.11 Furthermore, “the extent and scale of urban violence attributed to pandillas and maras is likely to be overstated” (Jütersonke et al. 2009: 380f). Most of the violence by both types of gangs instead seems “to be spatially circumscribed, isolated in large part to poorer peripheral communities from which the gangs emerge, and is also often inter-gang” (Ibid.: 381).

7Although it is outside the scope of this paper to establish whether or not the maras are linked to (transborder) organized crime, the above example of the role of the media highlights how issues are intentionally framed in a certain way. For example, Jütersonke et al. (2009: 380; emphasis added), point out that the federal or transnational character of the maras is “an imagined social morphology [rather] than a real phenomenon”. Edwards and Gill (2002) discuss two types of discourses related to TOC. One of these narratives, especially prevalent in the case of ministerial rhetoric, shows how governments “feel obliged to respond to popularly expressed insecurities or actually manipulate them for some broader political purpose” (Ibid.: 251f). Although increasingly discussed in recent years, symbolism in politics is certainly no new phenomenon. The “need for governing elites to provide reassurance that ‘something was being done’ about ‘problems’ ” (Edelman 1964, as referred to in Edwards & Gill 2002: 252) was already underscored by Edelman in 1964. Nowadays, this discourse is most apparent in the ambiguous language used in ‘wars’ against crime in general and especially against drugs (Edwards and Gill 2002: 252). The ‘war on gangs’ that characterizes the countries of the northern triangle in Central America is another such example and will be dealt with in detail in the following chapters.

2.3. The social and political construction of the maras in Central America’s northern triangle

8The Central American region faces some of the world’s highest homicide rates as well as some of the highest rates in reported criminal violence (see Table 1) (Ribando Seelke 2011), and the gang phenomenon can be considered to be “the most prominent aspect of the new landscape of Central American violence” (Rodgers et al. 2009: 1).12 However, levels of gang violence tend to be higher in the northern triangle (El Salvador, Honduras and Guatemala) than in Costa Rica, Nicaragua or Panama.13 Despite their prior existence before the civil wars in the region in the 1980s, the maras have increasingly received attention and become stigmatized during the past two decades.14 More precisely, they have become subject to sensationalist accounts on the part of politicians, the media and the general public. From a public opinion point of view, youth gangs often embody the “devil” responsible for all committed crimes.15 Isabel Aguilar Umaña (Director of the Central American Youth Program of Interpeace) highlights another form of stigmatization: the term ‘youth violence’ often directly creates the image of youths being the cause of violence, although they are “in many ways a symptom as well as a cause of a climate of insecurity that is overwhelming Central America” (Brevé 2007: 88; emphasis added).16 Aguilar Umaña therefore advocates ‘youth-related violence’ as a more balanced term. Moreover, she points out that the negative perceptions of youths go beyond Central America and that youths tend to be seen as a problem and not as a solution.17

9Actors outside the Central American isthmus have also asserted that youth gangs are an extraordinary security concern.18 For example in 2005, Anne Aguilera, then head of the Central America office at the International Narcotics and Law Enforcement Affairs unit of the U.S. Department of State, declared the maras were “the greatest problem for national security at [that] time in Central America and part of Mexico”.19 According to Boraz and Bruneau (2006), the maras not only threaten national security (such as the security of the state and the safeguarding of national sovereignty over both territory and population), but at the same time threaten public security (the maintenance of social order for the carrying out of commercial transactions and to provide transportation and communications services) and citizen security20 (the ability of individuals or groups to enjoy and/or exercise their economic, political and civil rights).21 In addition to posing threats to “all levels of security” (Boraz & Bruneau 2006: 39), the maras are also said to endanger economic development and democratic consolidation.

10The perception by citizens is a key factor of the mara phenomenon in Central America’s northern triangle. “Substantial levels of fear” (Desmond Arias 2011: 123) on the part of the population accompany the growing crime rates in the region and the so-called ‘fear of crime’ is present throughout Central America; be it in political debates or private conversations. The dominant opinion is of increases in violence and delinquency, as well as of an insecurity that is perceived to be higher than ever before (Peetz & Huhn 2008: 357f). For the case of El Salvador, Peetz (2011: 28 & 34) refers to Cohen’s (2011; first published in 1972) concept of ‘moral panic’ as regards the public’s fear of crime in general and youth gangs in particular. In the majority of the countries, public policies aimed at increasing citizen security have tended to be of a repressive nature, as shown by empirical data (Peetz & Huhn 2008: 352). Next to weakly implemented preventive approaches, between 2001 and 2006, “the official effort to confront gangs […] was oriented toward their suppression, and this policy dominated the political life of Central American societies while it was in force” (Cruz 2011: 145). Indeed, Peetz’ (2011: 1487) discourse analysis holds a “strong relationship and mutual influence between the public’s fear (or disregard) of youth violence and the state’s policies to reduce that kind of violence”. In this context, one must also take into consideration the ‘talk of crime’, which consists of “the permanent repetition and high social importance of the subject of insecurity in different spaces of communication (here in politics, the media and ‘everyday life’)” (Peetz & Huhn 2008: 364).22

11Peetz (2011: 1487) also suggests roads for further research, such as “to discover why, how, and by whom the discourses are originally generated” in order to facilitate the understanding of “the deeper roots of anti-youth crime policies that in some countries, such as El Salvador, tend to disregard human and children’s rights”. Although this paper is not based on newly-gathered empirical data or field research, looking at the mara phenomenon through the lens of securitization theory enables us to understand if the maras have been securitized, how these youth gangs are potentially socially and politically constructed, and in which context the resulting counter-policies are adopted. The theoretical foundations of securitization and critiques of the concept are the subjects to which I will turn in the following chapter.